Skip Navigation | ANU Home | Search ANU | Directories
The Australian National University
Research School of Earth Sciences
Printer Friendly Version of this Document
Untitled Document

The roles of bottom topography and internal gravity waves in horizontal convection

Ross W. Griffiths, Graham O. Hughes, Melissa A. Coman, Kial D. Stewart

Research School of Earth Sciences, Australian National University, Canberra, ACT 0200, Australia


Dye visualization of a horizontal convection experiment with bottom topography (the field of view includes only the left-hand half of the box). The overturning circulation is forced by surface cooling at the left-hand end of the box ("polar latitudes") and surface heating over the right-hand end ("equatorial latitudes"; not visible). The dye reveals that relatively warm surface waters flow from right to left above the sill, where they subsequently become very cold, sink, and are trapped on the left of the topography at levels below the sill. This trapped water eventually fills the left-hand basin and can be seen leaking across the top of the sill (from right to left), whence it enters the main basin as a highly localized sinking plume.

Horizontal convection is the flow generated by heating and cooling along one horizontal boundary of a box. Studies of this form of convection have yielded much insight into the fundamental dynamics governing the ocean overturning circulation. In particular, equatorial surface waters are heated and high latitude surface waters are cooled, and analogies with horizontal convection allow us to investigate the role of surface buoyancy forcing in the overturning circulation. We have examined the progress in this field in a major review article published this year (Hughes and Griffiths, 2008).

In 2008 we have built on our previous studies of horizontal convection, and work has progressed on two main fronts. Firstly, we have studied the role of bottom topography in restricting flows between ocean basins, and thus in controlling the rate of overturning circulation. Secondly, we have discovered the spontaneous generation of internal gravity waves in the convective flow.

Laboratory experiments (figure 1) have shown that the introduction of bottom topography blocks the circulation, isolating waters in adjoining basins below the level of the sill. The circulation is most strongly affected when the depth of the sill is comparable with the depth to which the surface thermal boundary layer extends (i.e. the thermocline). In this regime, the range of densities in the flow increases markedly and the rate of overturning decreases. On this basis we expect the Denmark Straits and Faroe Bank Channel in the North Atlantic and the Weddell Sea ice shelf overflow in the Southern Ocean to influence the global overturning circulation.

Numerical simulations of horizontal convection have revealed the existence of coherent propagating waves (figure 2). Further investigation has shown that highly localized sinking plumes (as in figure 1) perturb the density stratification and excite a spectrum of internal gravity waves. We have been able to identify interaction of these wave modes as a source of strong variability in the circulation. Further work is required to assess the importance of this phenomenon in the ocean overturning circulation.

Hovmöller plot of the vertical velocity along a horizontal section at mid-depth from a 2-D numerical simulation of horizontal convection. Time increases upwards; blue represents upwelling motion, green approximately no motion, and red downwelling motion. Regions of high latitude sinking are situated in this case at both the left and right hand ends of the horizontal section, and excite strong wave modes that propagate towards 'low latitudes' at the centre of the section. These waves and their interactions appear to be responsible for much of the variability in the circulation.

Hughes, G.O. and Griffiths, R.W. (2008) Horizontal convection. Annu. Rev. Fluid Mech. 40, 185-208.